Chapter 5
As the last chapter suggested, the Sântana-de-Mureş/Černjachov culture was the material expression of Gothic hegemony in the lower Danube region. That is to say, it was the product of a relative political stability that the imperial support for Gothic hegemony ensured. But the same stability held inherent dangers for the Roman empire. Constantine’s defeat of king Ariaric in 332, and the subsequent thirty years of peace between the empire and the Tervingi, did nothing to retard the growth of Tervingian military power. Thus when a Roman emperor next came to fight a major Gothic war, as Valens did in the later 360s, he confronted an opponent whose power would have surprised Constantine. More shocking still was the Gothic victory at Adrianople which, in 378, wiped out the larger part of the eastern Roman army. For us, it would be very satisfying to know just how Gothic power grew so great in this period. Unfortunately, we know remarkably little about the history of the Tervingian region in the three decades of peace that followed Constantine’s victory – not even whether we should talk about a Tervingian kingdom or kingdoms – and still less about more distant Gothic groups.
As we have already seen, it is impossible to be sure whether Ariaric was the only Tervingian king involved in the campaigns of 332, and how or whether he was related to later Tervingian rulers. The evidence for the mid fourth century is just as uncertain. In the year 364/365, we hear of more than one Gothic king sending troops to support an unsuccessful claimant to the imperial throne.[88] But in the later 360s and early 370s, our main narrative sources regard one particular Goth, Athanaric, as the sole leader of the Tervingi, even though they refer to him as iudex, ‘judge’, rather than as king. Other sources, however, demonstrate quite clearly that Athanaric was not the only Tervingian leader: we meet royal figures – a queen Gaatha, and others unnamed – who act in opposition to the iudex Athanaric, who is also called iudex regum, ‘judge of kings’, in one, admittedly highly rhetorical, source.[89] Thus, although a judge was clearly superior to a king, the substantive distinctions between them are altogether unclear, despite reams of scholarly speculation. Our Graeco-Roman sources were translating a genuine distinction between the Gothic word for king, reiks, a cognate of the Latin rex, and the Gothic word for judge, thiudans. The fact that thiudans was used to translate the Greekbasileús, ‘emperor’, in a Gothic martyrial calendar is a significant clue to the importance of the rank, but how much authority a Tervingian judge wielded over a Tervingian reiks, and how permanent it was, will always elude us.[90]
Constantine’s Death and Its Aftermath
Regardless, the short-term consequence of Constantine’s Gothic victory was merely to shift the focus of confrontation to a new set of barbarian enemies. In 334, Constantine campaigned against the Sarmatians, probably those who had asked for his help against the Goths in the first place. We are told that the servile population of the Sarmatian lands rebelled against their masters, and that many Sarmatians – 30,000 according to one source – fled into Roman service. Once inside Roman territory, they were divided among the Balkan and Italian provinces.[91] Whatever the shadowy events that had preceded the Gothic war, it is clear that the whole structure of the Danubian barbaricum had been deeply disturbed, as old hierarchies of power were overturned. After the 334 campaign, Constantine took the victory title Sarmaticus Maximus to accompany his multiple acclamations as Gothicus Maximus. He also took the title Dacicus Maximus, which probably represents a claim to have restored Trajan’s province of Dacia. The Carpathian lands of the old province were certainly not reannexed and subjected to Roman administration, but new garrisons in trans-Danubian quadriburgia and other small forts probably justified the claims. Nevertheless, Constantine was a familiar and frightening force beyond thelimes, as is illustrated by the large number of barbarian ambassadors present in 335 at the celebration of his tricennalia, his thirtieth anniversary on the throne, which is described to us by Eusebius, an eyewitness.[92]
The extent of Constantine’s prestige is illustrated by the immediate aftermath of his death in 337: for almost two years, we have no record of any campaigns against northern barbarians, an unheard of stability given that barbarian neighbours almost inevitably seized the opportunity provided by imperial successions to raid the Roman provinces. In the year of his death, Constantine had been preparing for a massive invasion of Persia, perhaps meant as the culmination of his world-conquering career. His death bequeathed a legacy of instability on the Persian frontier to his successors, and their own competition made matters worse. Constantine was succeeded by three sons and two nephews, the latter of whom died in a massacre of nearly all Constantine’s male relatives which was organized in order to ensure his sons a firm hold on the throne. But those sons – Constantinus, Constantius Ⅱ, and Constans – soon came to blows themselves. The eldest son, Constantinus, was displeased with his share in the division of the empire. He attacked his younger brother Constans in 340, but died in the war that followed. Thereafter, Constantius and Constans cohabited more or less peaceably for a decade, having probably campaigned together against the Sarmatians shortly before their elder brother’s death.[93]When Constans was overthrown by an army coup in 350, Constantius waged a bitter war against the usurper Magnentius. In the midst of this civil strife, the inhabitants of the frontier provinces were the greatest losers. Just as it had in the third century, the defence of the frontiers took second place to the prosecution of internal disputes, and thus the outbursts of civil war during the 340s and 350s encouraged barbarian incursions. Along the lower Danube, the peace of 332 continued to hold, the Tervingi honouring its terms and providing soldiers for the military ventures of the emperor Constantius. The Rhine frontier, by contrast, posed almost continuous difficulties. It was there that Constans had faced the revolt of Magnentius, and thence that Magnentius had drained troops in order to prosecute his war against Constantius. The connection between usurpation and barbarian invasion is made explicit in a speech attributed to the emperor Constantius, just before he appointed his cousin Julian as caesar and gave him the unwelcome task of restoring the Rhineland.[94]
Our Most Important Source: Ammianus Marcellinus
The source for this speech is Ammianus Marcellinus, one of the greatest historians of all antiquity and our main source for later fourth-century history, including the relationship between the Goths and the empire: without Ammianus, this and the next chapter could hardly be written. Given his importance to our understanding of the Goths, we must take a few moments to look at Ammianus himself before moving on. Ammianus was a Greek from one of the great cities of Syria, probably Antioch. He came from a good family which had been excused from its obligation to serve in the local town council and was instead closely linked to the larger imperial administration. As a young man, Ammianus served as protector domesticus, part of an elite group of soldiers who carried out a variety of special functions and often operated in the close vicinity of the emperor himself. Protectores frequently went on to command units of active-service troops in later life: the institution can be seen as a sort of officer training academy and more than oneprotector went on to become emperor. Ammianus found himself personally involved in a number of high-profile missions, and ultimately joined the invasion of Persia launched in 363 by the pagan emperor Julian, cousin and heir of Constantius. Unlike his Christian cousins and his uncle Constantine, Julian (r. 361–363) had repudiated Christianity, probably in reaction to the Christian piety of Constantine and his sons, who had murdered all but one of Julian’s close relations. Julian’s reign saw a determined attempt to undo Constantine’s Christianization of the empire, an attempt that fizzled out immediately upon Julian’s premature death on campaign. Julian, however, was Ammianus’ hero, and Ammianus may even have been an apostate from Christianity just as Julian was. Certainly his promising career came to a sudden halt with Julian’s death in 363 and it may be that Julian’s more committed pagan followers found their prospects stymied in the Christian reaction against the dead emperor.
Ammianus, his career prospects finished, devoted himself to research and, eventually, to writing the history of the Roman empire. We know that he travelled widely, and that he had moved from his native Greek East to Rome by 384. He wrote much of hishistory in Rome, perhaps under the patronage of one of the great senatorial families who dominated that city. His work liberally intermixes a political history of the empire with scholarly asides and Ammianus’ personal reminiscences. It was probably finished within a year or so of 390 and Ammianus may have died soon afterwards, for we know nothing more of him thereafter. The title he gave his history was Res Gestae, literally ‘deeds done’, and it indicates that the political history of the period it covers – from the reign of the emperor Nerva (r. 96–97) to that of Valens (r. 364–378) – was its primary concern. The surviving version of the text, unfortunately, begins with the start of book 14 and relates the events of summer 353 onwards, before ending, with book 31, just after the death of Valens at the battle of Adrianople in August 378. The structure of the history is complex and we cannot be sure of how it was originally organized, or indeed whether the numbering of the books as we have them is correct – it is possible that the original text consisted of thirty-six books, five of which are now lost and the rest misnumbered.[95] Almost all of Ammianus’ Gothic material is contained in the extant book 31, which is structurally and thematically very different from all those that precede it and may perhaps have been initially composed as a separate treatise. Regardless, the pessimism that pervades the Res Gestae is shaped by and focused on the catastrophic defeat of Valens by the Goths at Adrianople. The whole narrative is therefore filtered through the understanding of the terrible disaster that was to come, a disaster that Ammianus blames on a failure to honour Rome’s ancient traditions. This penetrating gloom must inform our own reading of Ammianus whenever we try to mine his text for information on the events of this period – we must always ask why he says what he says in the way he says it, for he is a master of innuendo and misdirection. On the other hand, he was a keen observer of Roman decline, and understood how Roman failures could lead directly to barbarian successes. Indeed, he alone demonstrates that at least some contemporaries understood and could articulate why the intersection of Roman internecine strife and barbarian invasion was so lethal: the barbarians ‘were like wild beasts who have acquired the habit of stealing their prey through the negligence of the shepherds’.[96]
Constantius on the Danube
The greatest barbarian danger, as the narrative of Ammianus shows, lay along the Rhine and the upper Danube, where neglect and the civil war that followed the death of Constans had weakened the frontiers. Both Alamanni and Franks were restive and the latter even succeeded in sacking so important a town as the imperial residence of Trier. Constantius, the last surviving heir of Constantine, was a deeply suspicious man, mistrustful of everyone, not least his own family. But he could not govern alone, certainly not with simultaneous disturbances on Rhine, Danube and eastern front, and even a suspect cousin was preferable to another usurper like Magnentius. Constantius turned to his only surviving male relatives, but the caesar Gallus, Julian’s elder brother, proved a disaster and soon met his end at the hands of the executioner. In 356, Julian alone remained and was duly appointed caesar. For half a decade, he campaigned more or less continuously along and beyond the Rhine. Constantius, meanwhile, devoted himself to the Sarmatians andQuadi of the middle Danube in year after year of campaigning. In 358 and 359, Constantius conducted massive punitive raids against Sarmatians and Quadi, and then against the Sarmatians’ former subjects the Limigantes. Deliberately sowing terror, Constantius sat in judgement on the many petty kings of the region, allocating territories to different groups.[97] As a result of these campaigns, the Sarmatians were eliminated as a serious power in the barbaricum. What is more, the suppression of the Limigantes created a sort of no-man’s land opposite the Danube bend between the powerful Quadic chieftains to the northwest and the Tervingian ones to the south and east. Both Quadi and Tervingi were to benefit.
Gothic power was presumably rendered all the more stable by these campaigns. Certainly neither Constantius, nor later Julian, ever felt the need to campaign along the ripa Gothica of the lower Danube, which was entirely peaceful between the 330s and the 360s. That peace allowed for the trade relations advertised in the name of a little fort called Commercium – ‘the marketplace’ – and for the recruitment of Gothic soldiers into the garrisons on the imperial side of the river.[98] Such garrison troops are probably responsible for the fairly widespread distribution of Sântana-de-Mureş/Černjachov decorative styles in places like Iatrus south of the frontier, and we know that Constantius was able to recruit many Goths for his Persian campaigns of the later 350s. The price of Constantius’ Danubian peace only became clear in the long term. His eventual successors, the imperial brothers Valentinian (r. 364–375) and Valens (r. 364–378), were spared any serious fighting between the Danube and the Tisza rivers where Constantius had suppressed the Sarmatians and Limigantes. But the absence of those troublesome neighbours only strengthened the power of Quadic and Tervingian rulers in their own territories and Valentinian and Valens each died on campaign, against the Quadi and the Goths respectively. Ammianus recounts how, in 361, the emperor Julian declared himself content to leave the Goths to the slave traders, so little did they merit military attention.[99] No doubt Ammianus wrote with the omniscience of hindsight, and wanted to bring home to us the tragic decline in Julian’s good judgement that would end in his death on a Persian battlefield. But if Julian really did speak those words or others like them, it was a stunning underestimate of Tervingian power. That power, in no small measure a product of imperial policy, would be revealed in three years of bitter warfare between the Tervingian iudex Athanaric and the emperor Valens. Before that, however, another element of imperial policy had begun to impinge heavily on Gothic society, in the shape of Christian missions.
Ulfila and Gothic Christianity
In the long years of stability after 332, Constantine’s ambition to evangelize the Goths was partly fulfilled. Constantine, as we have seen, had become a devoted Christian, certainly by the year 312 if not before. By the time he won his Gothic victory in 332, he had been implementing pro-Christian policies throughout the empire for several decades, particularly in the Greek East, which he had conquered from Licinius as the liberator of eastern Christians from persecution. He saw himself as a bishop to those outside the empire, and clearly regarded himself as called to evangelize the gentes beyond the frontier. The war against Persia which Constantine was preparing at the time of his death was prompted at least in part by his sense of Christian mission. Whether Constantine had explicitly planned for the evangelization of the Goths by 337 is controversial. It is sometimes argued that Constantine deliberately imposed Christianity on those Goths with whom he made peace in 332, but the evidence for that is not good.[100] On the contrary, it may simply have been a matter of chance that an opportunity to bring Christian teaching to the Goths arrived in the person of the Bishop Ulfila, sometimes known as Wulfila or Ulfilas.
Our information on the life of Ulfila is derived from just two sources, a letter written by one of his disciples, Auxentius, and a heavily abbreviated version of Philostorgius’ fifth-century Ecclesiastical History. Ulfila was descended from Cappadocians taken captive in the Gothic raids of Gallienus’ reign, but he himself bore a Gothic name. The date of his consecration as bishop and the start of his mission is debated. He came from Gothia on an embassy to the emperor – perhaps Constantine, perhaps Constantius Ⅱ – and was consecrated in either c. 336 or c. 341 by Eusebius of Nicomedia and other bishops. Eusebius was an adherent of a variety of Christianity associated with the Egyptian priest Arius who had argued that God the Son was subordinate to God the Father in the holy trinity. Arianism had been condemned as false doctrine – heresy – at the council of Nicaea, convened by Constantine in 325 immediately after his conquest of the eastern empire. In rejecting Arianism, the bishops at Nicaea decided that the Father and Son were identical, of the same substance (homoousion in Greek). Despite this, modified forms of Arius’ homoean theology – so-called from the Greek word for ‘likeness’, because it argued that the Father and Son were of like but not identical substance – continued to have considerable appeal, not least to Constantine, who was ultimately baptised by Eusebius of Nicomedia himself. Modern readers – including most professed Christians – are unused to and almost entirely indifferent to theology, and so find it very hard to understand why christological or trinitarian definitions aroused such passions in the early church. They did so because the consequence of getting the definition wrong – of believing the wrong thing about the persons of the trinity – was to compromise salvation. Because, after Constantine, the Roman state took on the task of guaranteeing Christian orthodoxy, the political cost of having one’s theological views condemned as heresy was very high. To understand the history of the fourth century, we need to take seriously both the political and the religious significance of theological disputes.
The fact that the Nicene position was not universally accepted by several fourth-century emperors was a particular complication. Constantius Ⅱ, for one, was a convinced supporter of homoean belief and attempted to enforce it as orthodoxy, even though he was defied by many bishops in the empire who supported the Nicene definition of the trinity. Ulfila was a member of the homoean party, and as a consequence of that fact, the earliest evangelism among the Goths brought Christianity in its homoean form. The homoean tendencies of Gothic Christianity were heavily re-enforced by later events under Valens, a fervent homoean who diligently persecuted his Nicene opponents. When he agreed to let many Tervingi into the empire in 376, he may have made some form of conversion a prerequisite for admission, and he had certainly sponsored missions among them before 376.[101] Valens, however, was the last emperor to support homoean doctrine and, upon his death, an overwhelming Nicene reaction meant that any doctrine with Arian tendencies would thereafter remain a reviled heresy. Over the years, however, Gothic Christians remained committed to their homoean doctrine and its homoean liturgy. When, a hundred years after Ulfila’s first mission, a powerful Gothic kingdom existed inside the Roman empire, Arianism functioned both as a defining symbol of Gothic identity, and as a major obstacle to peaceful coexistence between Gothic kings and the Nicene Romans over whom they ruled.
All the same, when his mission began, by 341 at the latest, Ulfila was simply adhering to the form of Christian doctrine endorsed as official orthodoxy by the emperor and those bishops whom he favoured. Ulfila was meant to serve as bishop for all the Christians already in the land of the Goths, but we have no idea how many such Christians there might have been, nor how many of them were descendants of former captives from the Roman empire and how many were converts won beyond the frontier. Within a decade of Ulfila’s arrival in Gothic territory, however, the number of Christians must have grown large enough to worry Gothic leaders, who associated the new religion with imperial power and therefore found the loyalty of Gothic Christians somewhat suspect. We do not know what sparked it, or indeed which Gothic leaders were involved, but eight years after Ulfila’s arrival in Gothia a persecution of Gothic Christians began. An offhand remark by bishop Cyril of Jerusalem seems to imply that this persecution produced martyrs, though none are known by name.[102] Ulfila and his followers were driven out of Gothic territory and into the empire, where they were granted lands in the province of Moesia Secunda, possibly around the city of Nicopolis.[103] Constantius addressed Ulfila as a new Moses, for leading his people out of servitude in their trans-Danubian Egypt.[104] Inside the empire, Ulfila became heavily involved in the ecclesiastical politics of Constantius’ reign and by the time of his death in 383 had been an influential theologian for many years.
The Gothic Bible
Ulfila and his followers in Moesia may have maintained close connections with co-religionists beyond the Danube, but we cannot be sure because the evidence comes from the fifth-century church historian Sozomen, who often misunderstands or oversimplifies fourth-century events. It would make good sense if Ulfila continued to be involved in diplomacy between emperors and Goths, yet in the 370s, the bishop of Tomi on the Black Sea, and not Ulfila, probably had responsibility for all the Christians of Scythia – both the Roman province of that name and the broader Gothic region beyond the frontier.[105] Regardless of that, Ulfila’s greatest impact on Gothic history came through his invention of an alphabet in which the Gothic language could be written. He based this alphabet on the Greek, but included new letters which could represent sounds not found in Greek.[106] Ulfila had only one purpose in creating this alphabet – to translate into Gothic the text of the Bible, so as to aid the work of evangelization. He translated into Gothic the whole text of the Bible apart, we are told, from the books of Kings, ‘because these books contain the history of wars, while the Gothic people, being lovers of war, were in need of something to restrain their passion for fighting rather than to incite them to it’.[107]This work of translation may well have involved not just Ulfila, but his followers as well, and was probably a product of their time in Moesia, rather than the eight short years they had been able to spend in Gothia. Yet the work they did endured. In the Gothic kingdoms of the fifth and sixth centuries, this Gothic Bible was the basic text for the homoean liturgy, and fragments of the Gothic Bible have been transmitted to us from many different sources. Almost all of these remains come from the New Testament, while only small fragments of Old Testament texts still survive. These biblical texts, however, are the earliest substantial evidence we possess for the morphology and vocabulary of a Germanic language, and are thus of priceless value to modern philologists.
Whether or not Ulfila’s mission was a direct product of Constantine’s own missionary ambitions, it was clearly a result of the Constantinian peace with the Tervingi. We have no way to correlate the growth of Christianity in Gothia with the meagre scraps of Tervingian history that are known to us during the reigns of Constantius and Julian. But we can be sure that Christianity was indeed spreading throughout the region, as retrospective evidence makes clear. As we shall soon see, the aftermath of Valens’ Gothic wars in the 360s brought on a second, much heavier, persecution of Christians in Gothic territory, one that is much better documented in ecclesiastical and liturgical sources. Most of the known victims of this second persecution seem to have been Nicene Goths, rather than homoeans. That would seem to imply that there were in fact two separate strands of missionary work beyond the lower Danube frontier in this period, however obscure their details may be to us.
Tervingi, Greuthungi and Other Goths
Still more obscure than the rise of Tervingian Christianity is the Gothic world beyond the Tervingi. There is no contemporary evidence, and almost everything we know about the larger Gothic world of the middle fourth century – apart from the archaeological evidence for its social structures, which we looked at in the last chapter – comes from retrospective accounts written after the disaster of Adrianople. Jordanes has much to say about this period, but it is almost all fiction that draws genuine figures from contemporary sources and inserts them into a spurious dynastic history of the sixth-century Ostrogothic king Theodoric the Great. The one thing we can be quite sure of is that beyond the fourth-century territory of the Tervingi there lay another Gothic realm, whose inhabitants were called Greuthungi. The Tervingi and Greuthungi have been interpreted as the linear ancestors of the later fifth-century Visigoths and Ostrogoths, and the long-standing division of the Goths into two sections under separate royal dynasties is a fixture of older literature (and still maintained by supporters of ethnogenesis-theory, with their insistence that royal dynasties transmit ethnic identity). In fact, the division between Visigoths and Ostrogoths is a product of fifth-century politics within the Roman and Hunnic empires, and the names are retrojection from the sixth-century text of Jordanes: they bear no demonstrable relationship to fourth-century divisions. The Gothic groups that emerged in the fourth century, after Adrianople, and in the fifth century, after the collapse of the Hunnic empire, were of thoroughly mixed origin, with connections to several different Gothic polities of the fourth century. Of these, the Tervingian polity is moderately well attested, but our knowledge of the Greuthungi comes almost entirely from a few pages of Ammianus Marcellinus. He, in turn, only mentions the Greuthungi when he describes the destruction of their kingdom by the Huns and the death of their king Ermanaric.
Ammianus actually knew very little about the Greuthungian kingdom. He tells us that Ermanaric ruled ‘lands rich and wide’ and was a ‘most warlike king and, on account of his many and various deeds, feared by the neighbouring peoples’.[108] That, in full, is the only contemporary evidence for Ermanaric’s kingdom that exists. Jordanes, however, expands that account into an elaborate list of peoples over whom Ermanaric held sway, drawing on traditions of classical ethnography and extending this fictional empire as far as northern Russia. Utter nonsense, Jordanes’ ‘empire of Ermanaric’ warrants none of the attention given it by otherwise serious scholars desperate for any scrap of information on the early Goths. Apart from the single line of Ammianus, the extent of Ermanaric’s power must remain a mystery to us. After his kingdom collapsed in the face of a Hunnic attack, we learn of several different groups of Greuthungi whom we cannot positively identify as having once been Ermanaric’s followers. That fact suggests that, just as different factions are known amongst the Tervingi in the face of Valens’ invasions of the 360s, so amongst the Greuthungi, Ermanaric was not the single source of power. The archaeological evidence we have looked at offers no help, and there is no material difference between the Sântana-de-Mureş/Černjachov territories in which the Tervingi were dominant and those in which Ermanaric’s Greuthungi lived. We must, in other words, content ourselves with only a very imperfect sense of Gothic history between the victory of Constantine in 332 and that of Valens in 369, to which we can now turn.
Valentinian and Valens
As with so much of the history of the Roman frontier, Valens’ Gothic wars are tied up in the internal conflicts of the empire, and particularly the legacy left to him by his predecessors Constantius Ⅱ and Julian. Julian, when we last saw him, had been appointed caesar by Constantius, in the hope that he would restore the Rhine frontier which had been so badly weakened by the usurpation of Magnentius. In 359, after many successes against Franks and Alamanni in the Rhineland, Julian was declared augustus by his troops.Both he and Constantius prepared for civil war, the latter bringing his Persian campaigns to an abrupt end in order to deal with his upstart cousin. A full-blown conflict was only averted by Constantius’ timely death, of natural causes, in 361. Julian immediately launched into a hugely ambitious program of reform, aimed both at reversing the Christianization of the Constantinian empire, and at fulfilling the dreams of his uncle and cousin and conquering Persia. After initial successes that brought the army to the walls of the Persian capital at Ctesiphon, Julian’s campaign ended in shambles; he himself died of a wound received in a sudden ambush. The army elected an ineffectual officer named Jovian (r. 363–364) to get them out of Persia, which he did at the cost of a humiliating peace-treaty that ceded several important Mesopotamian towns to the Persians. Jovian, a heavy drinker, soon died of self-indulgence, and the army high command elected Valentinian (r. 364–375) as emperor. Valentinian, a protector like the historian Ammianus, in turn appointed his younger brother Valens (r. 364–378) as his co-emperor.
The brothers were Pannonian, from the region of Lake Balaton in modern Hungary, and therefore from a region proverbially backward, the punchline in many a late Roman joke. This cultural stereotype deeply affects their treatment in our narrative sources, which invariably depict them as loathsome and cruel, a judgement mitigated only by Valentinian’s undeniable prowess as a general. Ammianus, the cultured Greek from Syria, found them unspeakable: Valens was subrusticum hominem, a half-witted man, who would have been murdered by his soldiers, had fate not spared him to suffer greater disaster.[109] Everything we know of Valens, in both Ammianus and the rich Greek historical tradition, is uniformly hostile. Our own interpretations of his reign, like those of his contemporaries, remain coloured not just by such slanders, but by his ultimate fate: killed by the Goths, along with most of the eastern Roman army, on the field of Adrianople. It has recently been argued that Valens was by no means a disastrous emperor, and certainly not the incompetent he is so often made out to have been. He was, on the contrary, a more or less average late Roman commander who faced an impossible concatenation of circumstances that ultimately brought him down.[110] Though that assessment may be a tritle generous, there can be no question that both Valens and Valentinian faced formidable challenges upon their accession to the throne.
The prospect of civil war between Julian and Constantius, and the fact that Julian launched his Persian war immediately upon the death of Constantius, led to the customary upheavals along the frontiers. As in the 340s and 350s, these disturbances were worst along the Rhine, though the Quadi, relieved of the pressure of their Sarmatian neighbours, required repeated campaigning by Valentinian to control. This demonstrates the structural dangers inherent in the standard imperial policy towards the barbarians.Constantius’ settlement of affairs between the Danube and Tisza had bred resentment amongst the Quadi, who had suffered demeaning punitive raids at the same time as the Sarmatians were suppressed, but it had also so strengthened them that as soon as the opportunity presented itself – as it did with Julian’s departure and death – they were able to launch devastating attacks on the provinces of Noricum and Valeria. So confident had the Quadi become in the security of their position that some of their envoys dared to address Valentinian as an equal during the campaigning season of 375. His outrage at this effrontery triggered the stroke which killed him, leaving the western provinces to his sons, one an untested youth named Gratian, the other, Gratian’s half-brother Valentinian Ⅱ, still only a toddler. Since 365, when they had divided the empire and its field army between them, the elder Valentinian and Valens had declined to interfere in one another’s affairs. Valens made no effort to intervene in the West upon his brother’s premature death, just as Valentinian had left Valens to his fate in the long series of disturbances that had faced him in the decade after 365. The earliest of these was the usurpation of Procopius, and it was from that venture that Valens’ Gothic campaigns ultimately stemmed.
The Usurpation of Procopius and the Breakdown of the Gothic Peace
Procopius could claim kinship with the Constantinian dynasty whose main line had died with Julian. He launched his usurpation at Constantinople in 365, suborning some troops who were en route to the Danube frontier, and almost succeeded in bringing down Valens’ new and insecure regime. Only the opportune treachery of some old associates of Constantius Ⅱ saved Valens, and Procopius was captured and executed in 366. Several Gothic kings had lent support to Procopius, supposedly sending 3,000 soldiers, but they excused themselves on the grounds of their treaty with the house of Constantine, whose legitimate heir they had taken Procopius to be.[111] We cannot tell if they believed their own excuses, but we do know their services were well rewarded: the largest hoard of silver coins in Gothic territory, from Caracal on the river Olt in modern Romania, contained nearly 3,000 silver coins, including thirty of Procopius.
Valens, as one can well imagine, had no intention of accepting this sort of excuse. He was in desperate need of a victory to shore up his prestige, badly damaged by a usurpation that had nearly unseated him. The Goths made an easier and more attractive target than did the intractable Persian frontier, and he could portray a Gothic war as well-merited punishment for lending support to a usurper. He seized the Goths who had come to support Procopius and deported them to Asia Minor. Then, in the three summer campaigning seasons from 367 to 369, Valens assaulted the Goths across the Danube. The campaigns were well prepared, as attested by a flurry of laws issued to the praetorian prefect Auxonius, responsible for organizing logistical support. What is more, the importance of the campaigns was widely anticipated in the eastern empire, for Valens received the dedication of a strange treatise, now anonymous, called De rebus bellicis, ‘On military matters’, which recommends both sensible measures well suited to Thracian conditions, and bizarre new war machines that no general could have deployed in reality. The orator Themistius, a great celebrity in Constantinople and a mouthpiece for imperial propaganda since the reign of Constantius, prepared public opinion for the successes that would soon be forthcoming. Sadly for Valens, Themistius’ enthusiasm went unvindicated by events.
Valens’ Three Gothic Campaigns
In the first campaign, launched in summer 367, Valens crossed the river at Daphne on a bridge of boats, which suggests that the Constantinian bridge from Oescus to Sucidava was no longer useable for large-scale military operations. The emperor laid waste the territory beyond the river, but failed to bring any large number of Goths to battle, because they fled into the Carpathians or the Transylvanian Alps in the face of his advance. He was, however, inspired to set a bounty on the heads of any Goth his men could capture, and this allowed him to at least salvage some claims to victory from the campaign.[112] In 368, rains and heavy flooding hampered the army’s movements, and Valens spent much if not all of the season encamped beside the Danube to no great military effect. He did, however, undertake a considerable construction campaign, restoring old and building new quadriburgia and smaller burgi, some of them named after himself or members of his family (for example Valentia, Valentiniana, Gratiana).[113] These building efforts are attested by bronze coins showing a burgus on their reverse, and by a fragmentary inscription from Cius that is dated to 368.[114] The third year of the war was more satisfying. After crossing the river at Noviodunum in the Dobrudja, Valens marched a long way into Gothic territory, sowing fear and destruction wherever he went. The Tervingian leader Athanaric gave battle and was defeated, as barbarian armies usually were when a Roman field army could pin them down in a set-piece battle. However, rather than pursue Athanaric in his retreat Valens returned to imperial territory, possibly because of the lateness of the season.[115]
The Terms of the Peace
Ammianus Marcellinus, who is our chief source for these campaigns, had reason to underplay their significance, knowing as he did that Adrianople was soon to come. But Valens’ three years of warfare had brought real successes. The new Danube forts strengthened Roman defences and the ability to project imperial power against the Goths. In fact, Valens shut down the frontier so effectively that Gothic access to Roman trade goods was systematically denied them. We saw in the last chapter how prominent a role trade with the Danube provinces played in the Sântana-de-Mureş/Černjachov regions, and Valens’ measures must have caused real hardship. More even than the battlefield defeat of 369, the shortages of Roman goods throughout Gothic territory forced Athanaric to sue for peace.[116] The terms of the peace were arranged in late summer 369 by two of Valens’ trusted generals, Victor and Arinthaeus. Valens and Athanaric met to solemnize a treaty near Noviodunum, but they did so on boats, midstream, Valens respecting the Gothic king’s oath not to set foot on Roman soil.[117] Hostages were given on the Gothic side, the emperor stopped paying subsidies to the Goths, and trade was opened up again, though restricted to just two (unnamed) cities.[118] Yet the separation of the Goths from the empire was not quite as complete as these measures were meant to suggest, for translators from Greek to Gothic continued to receive their imperial stipend, suggesting that the lines of communication were to be kept open.[119]
In the aftermath of this treaty, both sides could claim some sort of victory. The obliging Themistius, addressing the senate of Constantinople in early 370, preserves for us the official line: Valens’ philanthropy has inclined him to mercy. Why, after all, should a conquered and subjugated foe be wholly destroyed when he might be preserved and put to use on the battlefield? Valens, for his part, used the cessation of hostilities, and the concomitant propaganda triumph, to deal with growing trouble on the eastern frontier, taking up residence at Antioch in Syria for nearly half a decade. Athanaric, by extracting from the emperor a dignified peace on equal terms, was free to reassert his authority among the Tervingi. He chose to do this in part by launching a persecution of Gothic Christians, which may have led him into war against other Gothic chieftains and provoked further Roman intervention. Certainly, the opposition he experienced gives us some hint as to how low his prestige had fallen in three years of inconclusive warfare against Valens. As usual, the available sources leave much open to debate, and it is not at all clear that Gothic Christians had played any active role in helping Valens or opposing Athanaric before he began to persecute them. But, as had been the case with Diocletian decades earlier, suspicion might be grounds enough for persecution. Not only could Christians appear to poison the health of the state by refusing to honour its protecting deities, they were, from Athanaric’s point of view, potentially spies for the emperor. If, as seems quite possible, Ulfila’s Gothic community in Moesia maintained ties with co-religionists across the Danube, then Athanaric’s suspicions are thoroughly explicable. In times of peace, this sort of contact might be unproblematic, not much different than the to and fro of trade that so characterized the lower Danube of the mid fourth century. But once the Romans went to war against the Goths, and when Valens’ activities cut trade down to a tiny trickle, perspectives were necessarily altered. Gothic Christians might come to look less like fellow subjects of the Gothic kings and more like prospective sympathisers with Valens. If, as the fifth-century church historian Socrates tells us, Valens began to send missionaries into Gothia in 369, the case against Gothic Christians was that much clearer.[120] Persecution followed, its effects well documented in the extant sources.
The Story of Saba
The most extensive of these sources is the Passion of St. Saba, written shortly after 373, within a year or two of the death of its protagonist Saba. The Passion was sent to Basil, bishop of Caesarea in Cappadocia, perhaps the single most influential Greek bishop of his day.[121] Basil corresponded with the Cappadocian native Junius Soranus, who had been appointed to a military command in Roman Scythia – as dux Scythiae – in 373. This sort of letter exchange was a normal part of life for provincial elites, and the accession of a fellow provincial to an imperial office in a distant province usually meant an extension of patronage towards natives of the home region. Maintaining one’s network of correspondents was therefore an essential prerequisite of being able to serve one’s clients, and Basil’s letter collection is one of many that survive to show us how sedulously useful contacts were cultivated. The Passion of Saba is still available to us because of just this sort of letter exchange: a Balkan cleric (either Ascholius of Thessalonica or perhaps more likely a simple priest of the same name) seized the opportunity of Soranus’ appointment and his known connections to Basil to inform the great bishop of church affairs in the Danubian provinces and beyond. Basil, in gracious reply, flatters the author as a ‘trainer of Gothic martyrs’, though we have no way of evaluating the truth of that epithet. Regardless, the dux Soranus was enchanted by the legend of the holy Saba, sending men across the Danube to collect the saint’s relics and ultimately sending them toCappadocia, where they would thenceforth rest. The accidental connection of Gothic Christianity and Cappadocia, whence Ulfila’s ancestors had been taken more than a century earlier, was thereby perpetuated.
The story of Saba, as portrayed in his Passion, must detain us a little longer, for its incidental details offer the only glimpse of Gothic social history we possess apart from the archaeological remains of the Sântana-de-Mureş/Černjachov culture. Saba, we are told, was a villager somewhere in Gothia, perhaps in the region just to the southeast of the Carpathians. He was a Nicene rather than a homoean Christian, and may have been a cantor or lector in the local church (it is not altogether clear whether the reference to his “singing God’s praises in the church” should be given such a technical meaning). The Passion distinguishes several phases of persecution by the Gothic megistanes – ‘lords’ or ‘chiefs’, perhaps a direct reference to the Gothic king Rothesteus mentioned later in the text, perhaps to his more important followers. In both phases, these megistanes tested the loyalty of the villagers by forcing them to eat sacrificial meat. The first time this happened, the pagans in Saba’s village decided to trick the supervising officials by substituting meat that had not been sacrificed to the pagan gods for meat that had been. For us, this demonstrates the integration of Gothic Christians into village life and the willingness of their fellow villagers to unite against authority from outside the village, however legitimate it was.
For Greek contemporaries reading the Passion, however, it was Saba’s actions that proved his sanctity: refusing to go along with the deception, he made a conspicuous show of rejecting the meat altogether, and thus provoked his fellow villagers into exiling him from the village. He was allowed back before long, but promptly stirred up further trouble for himself and the other Christians of the village. When a Gothic noble came to the village for a second time to supervise the consumption of the sacrificial meat, the pagan villagers were going to swear, while eating it, that there were no Christians in the village. Once again, Saba revealed himself and refused to play along. But when the villagers swore that Saba was a man of no account, possessing ‘nothing but the clothes he wears’, the Gothic lord did no more than order his expulsion from the gathering, on the grounds that a man with no property could neither help nor harm. That response is strong evidence for the essentially political nature of the persecution in Gothic territory: powerful Gothic converts might be a threat, potentially in league with the emperor; a man like Saba was at worst a conspicuous nuisance.
Yet in the final phase of the persecution, Saba’s obstinacy reached a pitch that provoked the martyrdom he so clearly craved. Saba was en route to another village to celebrate Easter with a priest named Gouththikas, when a miraculous fall of snow prevented his going forward and turned him back to celebrate the feast in his own village with his fellow Christian, the priest Sansalas. Three days after Easter, Atharidus, the son of the Gothic king Rothesteus, arrived in the village with an entourage, specifically to arrest Sansalas. Saba, found in his company, was likewise arrested, but while Sansalas was held captive to face a higher authority, Saba was tortured on the spot. First driven through thorny thickets, then lashed to the axles of a wagon and flogged through the night, he defied his tormentors in the approved manner of one born to martyrdom. A friendly village servant freed him and fed him, but the tortures continued on the following day, when Sansalas and Saba were ordered by Atharidus’ men to consume meat that had been sacrificed. Saba naturally refused and was finally condemned to die, on 12 April 372. The soldiers chosen to drown him in the river Musaios, perhaps the Buzaǔ, seriously considered him setting him free: they thought him simple-minded for rejoicing at his coming martyrdom and reasoned that Atharidus would never find out if they just let him go. But Saba, insisting that he could see an army of saints waiting beyond the river to welcome him into heaven, urged them to their duty. So ‘they took him down to the water, threw him in and, pressing a beam against his neck, pushed him to the bottom and held him there’. It may have been Sansalas who set down in writing the account of Saba’s martyrdom.[122]
Other Gothic Martyrs and the Motives for Persecution
Saba was not the only martyr in this persecution. As St. Jerome put it in his chronicle entry for 369, ‘Aithanaric king of the Goths persecuted Christians, killed many and drove them from their own lands to the lands of the empire’.[123] Quite a few of these martyrs are recorded by name, in several different sources. Of particular significance is the list of martyrs remembered at Cyzicus in Asia Minor, where relics were deposited by Dulcilla, the daughter of an otherwise unknown Gothic queen named Gaatha. Among these martyrs we find the priests Bathouses and Wereka, their unnamed children, the monk Arpulas, eleven named Gothic men, and seven named Gothic women, all killed at the command of the Gothic leader Wiguric.[124] Other names are known from less reliable sources, but the drift of the evidence is clear: while some Gothic leaders favoured Christianity and tried to preserve the memory of their local martyrs, many also supported Athanaric’s persecution.
We have already seen why Athanaric might rationally have regarded Gothic Christianity as a threat that needed to be stamped out. Yet in the face of this essentially political explanation, it is worth pointing out that some of his followers will have supported him out of genuine conviction. The story of Saba makes that clear: in the first of his several confrontations with Gothic authorities, Saba was exhorted to eat the sacrificial meat in order to save his soul. Unless this is merely a Christian gloss put on the confrontation by an ecclesiastical author, it would seem that some Goths regarded Christianity not just as a threat to the Gothic state, but also to the spiritual health of converts. All the same, it would be hard to deny that political fears were the foremost motive for persecution. The enthusiasm of leaders like Rothesteus and Wiguric shows that well below the level of the iudex Athanaric, it was feared that Christians might form a fifth column more sympathetic to the empire than to pagan Gothic leaders.
That some members of the Gothic aristocracy had converted only made matters worse, for they were in a position to treat with the Roman empire where a man like Saba was not. Emperors were only too keen to encourage dissension among barbarian neighbours, and Athanaric’s defeat had damaged his authority, however much face he saved by the peaceful compromise that ended the war. As we have seen, the church historian Socrates reports that Valens used the peace to evangelize the Goths. Socrates also reports that the Tervingian chief Fritigern was one of those converted, that his Christianity caused him to go to war with Athanaric, and that some Roman soldiers were sent to aid Fritigern before the two Gothic leaders made peace. The conversion story is corroborated by Fritigern’s probable commemoration in a later Gothic liturgical calendar, though the Gothic civil war is known only from Socrates.[125] Yet it too is plausible. The next time we meet Fritigern – in Ammianus, a more reliable source than Socrates – he is the leader of Tervingi opposed to Athanaric. We cannot know whether Fritigern opposed Athanaric because he was a Christian, or whether he became a Christian because he opposed Athanaric. But it seems quite clear that Athanaric was absolutely right to see the extension of Christianity among Gothic elites as a substantial threat to the political status quo in Gothia. How matters might have turned out in the long run is a moot point. Within four years of Saba’s martyrdom the stability of the whole Gothic world had been shattered by obscure but traumatic events that brought many Tervingi to the banks of the Danube, begging for admission to the empire, in the spring of 376.